1 / 21

Conversation Level Constraints on Pedophile Detection in Chat Rooms

Conversation Level Constraints on Pedophile Detection in Chat Rooms. PAN 2012 — Sexual Predator Identification. Claudia Peersman, Frederik Vaassen , Vincent Van Asch and Walter Daelemans. Overview. Task 1: Sexual Predator Identification Preprocessing Experimental Setup and Results

andra
Download Presentation

Conversation Level Constraints on Pedophile Detection in Chat Rooms

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Conversation Level Constraints on Pedophile Detection in Chat Rooms PAN 2012 — Sexual Predator Identification Claudia Peersman, FrederikVaassen, Vincent Van Asch and Walter Daelemans

  2. Overview • Task 1: Sexual Predator Identification • Preprocessing • Experimental Setup and Results • Test Run Results • Task 2: Identifying Grooming Posts • Grooming Dictionary • Test Run Results • Discussion

  3. Task 1: Preprocessing of the Data • Data: PAN 2012 competition training set • predator vs. non-predator • info on the conversation, user and post level • Two splits: training and validation set • No user was present in more than one cluster  prevent overfitting of user-specific features

  4. Experimental Setup • Features: token unigrams • LiBSVM • Probability output • Parameter optimization • Experiments on 3 levels • data resampling

  5. Level 1: the Post Classifier • Resample the number of posts  Equal distribution of posts per class • About 40,000 posts per class in training • No resampling in the validation sets

  6. Level 1: the Post Classifier (2) • Only output on the post level • Aggregate the post level predictions to the user level: • LiBSVM’s probability outputs • Predators= average of the 10 highest predator class probabilities ≥ 0.85

  7. Results for the Predator Class

  8. Level 2: the User Classifier • Resampling on the user level  exclude users with no suspicious posts • Filter: dictionary of grooming vocabulary  see Task 2 • Why? • reduce the amount of data • “hard” classification  higher precision?

  9. Update Results (1) Data reduction: up to 48.4%  Combine systems?

  10. Combining the systems • Weighted voting using LiBSVM’s probability outputs • 70% of the weight on the high precision User Classifier

  11. Update Results (2)

  12. Level 3: Conversation Level Constraints • Both users in a conversation labeled as predators • Our approach: • go back to predator probability output • use the high precision user classifier • Predator probability ≥ 0.75

  13. System Overview Post Classifier User Classifier

  14. Update Results (3)

  15. Results on the PAN 2012 Test Set • Future research: • more splits • investigate ensembles

  16. Task 2: Identifying Grooming Posts • From the final predator ID list  detect posts expressing typical grooming behavior • No gold standard labels  What is grooming? • Predator conversations have predictable stages (e.g. Lanning, 2010; McGhee et al., 2011)

  17. Task 2: Identifying Grooming Posts (2) • Dictionary containing references to 6 stages: • sexual topic • reframing • approach • data requests • isolation from adult supervision • age (difference)

  18. Task 2: Identifying Grooming Posts (3) • Resources: • McGhee et al. (2011) • English Urban Dictionary websitehttp://www.urbandictionary.com/ • English Synonymshttp://www.synonym.net/ • cf. user classifier filter

  19. Results on the PAN 2012 Test Set • Precision = 0.36 • Recall = 0.26 • F-score (β = 1) = 0.30

  20. Discussion • Use of β-factors to calculate the F-score: • Task 1: focus on precision (β = 0.5) • Task 2: focus on recall (β = 3.0) • However, in practice: • find all predators (recall in Task 1) • find the most striking posts (precision in Task 2)

  21. Questions? Contact: claudia.peersman@ua.ac.be

More Related